You will lead research on the models behind Natural General Intelligence. Open questions include how to represent the 3D subsurface and its history, how to learn from a few hundred labelled deposits alongside terabytes of unlabelled data and millions of simulations, and how to tell whether a prediction is right where nobody has drilled.
What your first year looks like
- Lead research in foundation models for Earth observation and drill core, generative 3D geology, or world models.
- Develop methods that combine simulation, sparse field data and large unlabelled datasets.
- Define benchmarks and calibration tests that reflect real exploration decisions.
- Publish where it helps, and mentor engineers and interns.
You have:
- A PhD in ML, statistics, physics or applied mathematics.
- First-author publications at NeurIPS, ICML, ICLR or leading scientific journals.
- Depth in generative modelling, self-supervised learning, probabilistic inference or scientific ML.
- The engineering skill to turn ideas into working systems.
Nice to have:
- ML applied to a physical science.
- Large-scale training on scientific or geospatial data.
Interested? Write to us at gondwana@altcarbon.com with the role in the subject line.